Software of Unknown Provenance
SOUP. It is likely that you are familiar with the acronym, SOUP, in relation to medical device and Health IT software. The medical device software standard IEC 62304, defines SOUP as a “software item that is:
- already developed and generally available and that has not been developed for the purpose of being incorporated into the medical device (also known as “off-the-shelf software”) or
- software previously developed for which adequate records of the development processes are not available.
When I started my career in the mid-1980’s, we developed most ALL of the software in our medical devices. Very little SOUP was used. However, in 2021, most medical devices and Health IT systems are composed primarily of SOUP! Wow, what a transition!
We are often asked about SOUP from a compliance standpoint. One of the major confusions is confusing SOUP used in our products (which are subject to IEC 62304) with SOUP that may be used in our development and other quality system tools (which are not subject to IEC 62304). IEC 62304 is concerned with the quality of the software running in your medical devices (including SaMD).
A second confusion is artificially gather SOUP documentation separately then where it would naturally apply. For example, having a separate SOUP software risk analysis from the custom software risk analysis. Likely, SOUP failure modes are in the same fault trees as your custom software so it would be easiest for the software development team to do the software risk analysis of custom software and SOUP at the same time and document it in the same way. This applies to all places where SOUP should be considered in the entire SDLC.
Don’t create unnecessary work simply because you think that it is required for compliance. That is not the intent of IEC 62304 nor of the US design controls regulations. Sometimes we may see these separate documentation tasks as improving quality – but in fact, they could have the opposite effect! The time invested by the software development teams creating unnecessary documentation could have in fact been invested in greater rigor and detail for safety related software.
Gotta use SOUP? I get it – but analyze its failure modes and design in risk controls. We will all benefit from it. Onward!